Карасев Александр Владимирович : другие произведения.

Measurement problem in the neural picture of the world

Самиздат: [Регистрация] [Найти] [Рейтинги] [Обсуждения] [Новинки] [Обзоры] [Помощь|Техвопросы]
Ссылки:
 Ваша оценка:
  • Аннотация:
    In the atomistic picture of the world, the measurement problem is divided into three fundamentally different, in no way related processes. In the neural picture of the world, these processes are represented by successive stages of a single holistic algorithm for changing the state of the neural network. From this point of view, many classical paradoxes do not look so paradoxical.
  Measurement problem in the neural picture of the world
  
   A. V. KARASEV
  
   In the atomistic picture of the world, the measurement problem is divided into three fundamentally different, in no way related processes.
   In the neural picture of the world, these processes are represented by successive stages of a single holistic algorithm for changing the state of the neural network.
   From this point of view, many classical paradoxes do not look so paradoxical.
  
  In quantum mechanics, the observation problem is divided into three processes [1].
   Process 1 is the observer's decision as to what question he will ask the quantum world.
   Process 2 - evolution of a state according to the scheme described by the Schrödinger equation.
   Process 3 is a quantum state, which is the answer to the question formulated during the implementation of process 1, or reduction of the state.
  In the traditional - atomistic picture of the world, all these processes are fundamentally different from each other. They differ so much that no unity is visible between them. The participants in these processes - the investigated object, the measuring device and the observer - are also fundamentally different. The object under study is described by quantum mechanics. But the measuring device, on the contrary, is always assumed to be a fundamentally classical object.
   And the observer does seem to be an alien creature, fundamentally excluded from the picture of the universe [2].
   Let's consider how these processes look when translated into neural terminology [3]. Any look at the problem from a new point of view can reveal something unexpected and promising.
  To begin with, in the neural picture of the world, all three participants - an object, a device, and an internal observer - are represented as separate fragments of a single neural network of the Universe. And already in this lies the organic unity and integrity of the neural picture of the world in comparison with the atomistic one. Of course, a fragment that models an internal observer will be somewhat more complex than a fragment that represents a physical object. But not much more complicated.
  For adequate observation of physical objects, an internal observer needs only associative memory. And neural networks may well possess this property [3].
   Process 1 means that the internal observer of the neural network creates a special piece of the neural network - a measuring device. The source of particles for this device will be represented as an input neural connection. Ideally, this connection should be the only one - after all, in physics we want to study a certain object by itself, in isolation from the rest of the world. That is, a fragment of a neural network representing a measuring device must have a single input thread from the rest of the neural network of the Universe.
  The inner observer itself does not create neural connections. He only uses ready-made fragments of a neural network in the form of sufficiently rigid and sufficiently strong bodies already existing in nature. For example, a blank screen is such a fragment of a neural network created by nature itself. The set of inputs of this fragment-screen is presented to the observer in the form of some opaque surface. Opacity means that input signals arriving at this surface are lost in the labyrinth of internal connections of the fragment.
  If the observer has sufficiently rigid bodies, then it will be convenient for him to construct the Euclidean geometry - it is convenient to the extent that the approximation of the hardness of these bodies works.
   If the observer does not have enough solid bodies - for example, on a cosmological or, conversely, a microscopic scale - it will be more convenient for him to choose the maximum speed of transmission of neural signals as a standard of space-time and, therefore, he will use the Lorentz geometry [4].
  The observer receives all information about a quantum object from counters of various particles. In neural terminology, such counters are represented by such fragments of a neural network, in which the minimum input signal is branched and multiplied into a sufficiently powerful output signal, which is observed at the macroscopic level in the form of, for example, the click of a counter or the movement of an arrow on a measuring scale.
  All these prepared (both by nature and by the observer) fragments of the neural network are combined by the observer into such a measuring device, which should give an answer to the question posed in Process 1. It is obvious that the designs of devices for different measurements can be fundamentally different. Therefore, in neural terminology, there is nothing surprising in the fact that, for example, a coordinate and an impulse cannot be measured simultaneously. Indeed, to measure them, it is necessary to organize fundamentally different matrices of neural connections and streams of neural signals.
  Finally the measuring device is ready and Process 2 begins. The observer sends some neural signal to the input of this device. This signal can branch and multiply - depending on the specific organization of the neural connections that make up this device [5-6]. In this case, the evolution of a neural signal is described by the Schrödinger equation in neural terminology.
  
   V(i) = SUM( j) T(ij) U(j)
  
  where j is the address of the first neuron, U(j) is the signal at its output;
   i - the address of the second neuron, at the input of which the signal V(i) is summed;
   T(ij) is a matrix of communication between neurons from j to i, created as a result of process 1.
  The signals U, V are analogous to the amplitude of the probability of an event - the excitation of the particle counter with number i. The connection between neurons T(ij) is similar to the amplitude of the transition . The values T, U, V are of course complex, which is provided by the additional graph Im (imaginary unit) in the tables of neural signals and connections [5-6].
  As in traditional atomistic terminology, process 2 does not cause any fundamental problems - it is just a continuous evolution of the wave function in the absence of events (measurements).
  Fundamental problems begin with process 3, in which the wave function (state vector) changes abruptly. In traditional terminology, this process is the most difficult to understand [7]. How is this so - all our mathematics is based on the analysis of continuous smooth functions - and suddenly gap, jump. Especially in the case of paradox EPR.
  In neural terminology, process 3 looks much more natural, since the dynamics of the state vector (neural signals) is described here not by operators, but by neural network algorithms. And the command to change the state of all neurons in the network or one, separately taken fragment of it is just one of many similar algorithms. The scope of this command determines the degree of coherence of the network fragment. That is, if some neurons can change their states with one command, at one clock of the neurocomputer, then it is all these neurons that make up a coherent fragment of the network, which appears to the inner observer as a coherent physical object.
   It does not matter how much coherent neurons are spatially separated from each other. That is, how many intermediate neurons separate them (or can be inserted between them). All the same, the scope of the state change command does not change - it will pass through all the neurons that were coherent from the beginning. Therefore, EPR no longer looks like a paradox [6].
  So, in neural terminology, all three measurement processes look like simply three stages of a single integral algorithm for changing the state of a neural network, in which both the object under study and the observer with a measuring device participate organically. All of them are represented by fragments of a single Universe neural network with uniform algorithms for changing states. For the study of physical objects, these algorithms are as simple as possible. But here a natural question inevitably arises - are there more complex, but essentially similar algorithms for changing the state in the same network? However, this is a completely different story, far beyond the scope of physics [8].
  
  Literature
  
   1. William Arntz, Betsy Chase, Mark Vicente "What do we even know?"
   2. Schrödinger E. Nature and the Greeks
   3. Karasev A. V. Neural picture of the world. Bulletin of new medical technologies. 2002.vol. 9.N 2. http: //samlib.ru/k/karasew_a_w/nkmfs.shtml
   4. Karasev A.V. Observer model in Poincaré's picture of the world.
   http://samlib.ru/k/karasew_a_w/poincare.shtml
   5. Karasev A.V. Three-dimensional space and electron spin in neural terminology. Quantum Magic, 2011, volume 8, issue. 2.http: //quantmagic.narod.ru/volumes/VOL822011/p2168.pdf
   6. Karasev A.V. The EPR paradox in the neural terminology of quantum mechanics. http://samlib.ru/k/karasew_a_w/eprfs.shtml
   7. This is written in all the books.
   8. Karasev A.V. Than a neural picture of the world different from speculation about - we live in a computer. http://samlib.ru/k/karasew_a_w/nkm_diff.shtml
  
 Ваша оценка:
Связаться с программистом сайта.

Новые книги авторов СИ, вышедшие из печати:
О.Болдырева "Крадуш. Чужие души" М.Николаев "Вторжение на Землю"

Как попасть в этoт список
Сайт - "Художники" .. || .. Доска об'явлений "Книги"